CN113747851A - Automated electrode recommendation for ablation systems - Google Patents

Abstract

An ablation system comprising: a Radio Frequency (RF) generator configured to generate RF energy; an ablation catheter in communication with the RF generator and including a plurality of ablation electrodes; a camera positioned on the ablation catheter and arranged to take an image including at least some of the plurality of ablation electrodes; and one or more processors configured to recommend a subset of the plurality of ablation electrodes to be active ablation electrodes based at least in part on the image.

Description

Automated electrode recommendation for ablation systems

Cross Reference to Related Applications

This application claims priority to provisional patent application No. 62/822,142 filed on 22/3/2019, the entire contents of which are incorporated herein by reference.

Technical Field

The present disclosure relates generally to systems, devices, and methods related to cardiac ablation.

Background

Various cardiac abnormalities may be attributed to inappropriate electrical activity of cardiac tissue. Such inappropriate electrical activity may include generating electrical signals, conducting electrical signals of tissue, etc. in a manner that does not support effective and/or efficient cardiac function. For example, a region of cardiac tissue may be electrically active prematurely or otherwise asynchronously during the cardiac cycle, thereby causing the cardiomyocytes in that region and/or adjacent regions to contract out of rhythm. The result is abnormal cardiac contraction that does not achieve optimal cardiac output. In some cases, a region of cardiac tissue may provide a false electrical path (e.g., a short circuit) that leads to an arrhythmia, such as atrial fibrillation or supraventricular tachycardia. In some cases, non-viable tissue (e.g., scar tissue) may be preferred over malfunctioning heart tissue.

Disclosure of Invention

In example 1, a computer-implemented method includes recommending, via one or more processors and a graphical user interface, a set of ablation electrodes to activate based at least in part on an analysis of an image containing the ablation electrodes.

In example 2, the method of example 1, wherein analyzing comprises identifying an ablation electrode in the image.

In example 3, the method of any one of examples 1 or 2, wherein analyzing includes comparing contrast of pixels in the image around the ablation electrode.

In example 4, the method of any of examples 1 to 3, wherein the analyzing determines which of the ablation electrodes are selectable.

In example 5, the method of any of examples 1 to 4, wherein the analysis is performed by a trained neural network.

In example 6, the method of any of examples 1-5, wherein the recommendation is also based at least in part on the impedance measurement.

In example 7, the method of example 6, wherein the impedance measurements determine which of the ablation electrodes are selectable.

In example 8, the method of example 7, wherein the selectable ablation electrodes are ablation electrodes associated with impedance measurements within a predetermined range of impedance values.

In example 9, the method according to any one of examples 4, 5, 7, and 8, further comprising assigning, via the one or more processors, each of the recommended ablation electrodes as a sink or a source.

In example 10, the method of any of examples 1 to 9, wherein the recommending balances based at least in part on an estimated power to be received by the source electrode.

In example 11, the method of any of examples 1 to 10, wherein the recommendation is based at least in part on limiting the estimated power received by the source electrode to below a predetermined threshold.

In example 12, the method of any of examples 1-11, wherein the recommended set of ablation electrodes forms a closed circuit.

In example 13, a computing device adapted to perform the steps of the methods of examples 1 to 12.

In example 14, a computer program product comprising instructions to cause one or more processors to perform the steps of the method of examples 1 to 12.

In example 15, a computer-readable medium has the computer program product of example 14 stored thereon.

In example 16, an ablation system includes: a Radio Frequency (RF) generator configured to generate RF energy; an ablation catheter in communication with the RF generator and including a plurality of ablation electrodes; a camera positioned on the ablation catheter and arranged to take an image including at least some of the plurality of ablation electrodes; and one or more processors configured to recommend a subset of the plurality of ablation electrodes to be active ablation electrodes based at least in part on the image.

In example 17, the ablation system of example 16, wherein the recommendation is based at least in part on comparing contrasts of pixels in the image around the ablation electrode.

In example 18, the ablation system according to any of examples 16 and 17, wherein the one or more processors are configured to determine which of the ablation electrodes are selectable based at least in part on a comparison of pixels in the image.

In example 19, the ablation system of any of examples 16-18, wherein the recommendation is also based at least in part on impedance measurements made by the plurality of ablation electrodes.

In example 20, the ablation system of any of examples 16 to 19, wherein the recommended ablation electrode is an ablation electrode associated with an impedance measurement within a predetermined range of impedance values.

In example 21, the ablation system of any of examples 16-20, wherein only some of the plurality of ablation electrodes are selectable, wherein the one or more processors are configured to assign each of the selected ablation electrodes as a sink or a source.

In example 22, the ablation system of any of examples 16 to 21, wherein the recommendation is based at least in part on limiting an estimated power received by the sink electrode to below a predetermined threshold.

In example 23, the ablation system of any of examples 16 to 22, wherein the recommended ablation electrode forms a closed circuit.

In example 24, the ablation system of any of examples 16 to 23, wherein the recommendation is based at least in part on images of the ablation electrode taken over a plurality of cardiac cycles.

In example 25, the ablation system of any of examples 16 to 24, wherein the ablation catheter comprises an expandable member carrying a plurality of ablation electrodes, wherein the camera is positioned within the expandable member.

In example 26, a computing device for generating and using a Graphical User Interface (GUI) is disclosed. The computing device includes one or more integrated circuits configured to generate, for display via the GUI, graphical representations of a plurality of electrodes of the ablation catheter, and automatically highlight, on the GUI, a subset of the plurality of electrodes as active electrodes based at least in part on an impedance value associated with each of the plurality of electrodes.

In example 27, the computing device of example 26, wherein the automatically highlighting is further based at least in part on an analysis of an image of an electrode comprising the ablation catheter.

In example 28, the computing device of example 27, wherein analyzing comprises comparing contrast of pixels in the image around the electrode.

In example 29, the computing device of any one of examples 27 and 28, wherein the analyzing comprises identifying each of the electrodes in the image.

In example 30, the computing device of any of examples 27 to 29, wherein the analyzing is performed at least in part by a trained neural network.

In example 31, the computing device of any of examples 26 to 30, wherein the one or more integrated circuits are further configured to automatically designate each of the highlighted subset of the plurality of electrodes as a source electrode or a sink electrode.

In example 32, the computing device of example 31, wherein the one or more integrated circuits are further configured to automatically dispense an amount of energy to each of the source electrodes.

In example 33, the computing device of any of examples 31 and 32, wherein the amount of energy allocated is based at least in part on an estimated power calculated for each of the designated sink electrodes.

In example 34, the computing device of example 33, wherein the amount of energy allocated is based at least in part on balancing the calculated estimated power between the designated sink electrodes.

In example 35, the computing device of any of examples 26 to 34, wherein automatically highlighting is further based at least in part on determining which of the electrodes of the ablation catheter are selectable.

While multiple embodiments are disclosed, still other embodiments of the present invention will become readily apparent to those skilled in the art from the following detailed description, which shows and describes illustrative embodiments of the invention. Accordingly, the drawings and detailed description are to be regarded as illustrative in nature and not as restrictive.

Drawings

Fig. 1 illustrates an ablation system according to certain embodiments of the present disclosure.

Fig. 2 illustrates a perspective view of an ablation catheter in accordance with certain embodiments of the present disclosure.

Fig. 3 and 4 illustrate various views of a graphical user interface according to certain embodiments of the present disclosure.

Fig. 5 shows a schematic representation of an electrode selection process according to certain embodiments of the present disclosure.

Fig. 6A-6C illustrate images captured by a camera of the ablation catheter of fig. 2, according to certain embodiments of the present disclosure.

Fig. 7 shows a schematic representation of features of a neural network, in accordance with certain embodiments of the present disclosure.

Fig. 8 illustrates a block diagram representation of steps in a method of recommending an ablation path, in accordance with certain embodiments of the present disclosure.

Fig. 9 illustrates a schematic representation of an electrode of the ablation catheter of fig. 2, in accordance with certain embodiments of the present disclosure.

While the invention is amenable to various modifications and alternative forms, specifics thereof have been shown by way of example in the drawings and will be described in detail. However, the intention is not to limit the invention to the particular embodiments described. On the contrary, the intention is to cover all modifications, equivalents, and alternatives falling within the scope of the invention as defined by the appended claims.

Detailed Description

Cardiac ablation is a procedure in which cardiac tissue is treated to inactivate the tissue. Tissue targeted for ablation may be associated with inappropriate electrical activity, as described above. Cardiac ablation can create lesions in tissue and prevent the tissue from improperly generating or conducting electrical signals. For example, damage in the form of a line or circle may prevent propagation of a false electrical signal. It is desirable to control the shape, depth, uniformity, etc. of the lesion.

Certain embodiments of the present disclosure relate to systems, devices, and methods that may be used in conjunction with cardiac ablation therapy via an ablation electrode on an ablation catheter. In particular, the present disclosure describes methods for recommending and/or selecting which ablation electrodes to activate for treatment. The recommended ablation path helps to enhance the process of determining which electrodes to activate during an ablation procedure. Further, the present disclosure describes a graphical user interface that displays and enables control of a graphical representation of a feature of an ablation catheter, and that may be used to view, select, and modify ablation parameters, and the like.

Fig. 1 shows an ablation system 100 including an ablation catheter 102, the ablation catheter 102 including an elongate catheter body 104 and a distal catheter region 106, the distal catheter region 106 configured to be positioned within a heart 108. The ablation catheter 102 includes an expandable member 110 (e.g., a membrane, a balloon) and a plurality of energy delivery elements 112 (e.g., ablation electrodes) secured to the expandable member 110. The energy delivery element 112 is configured and positioned to deliver ablative energy (e.g., Radio Frequency (RF) energy) to tissue when the expandable member 110 is expanded.

The system 100 includes an RF generator 114, the RF generator 114 electrically coupled to the plurality of energy delivery elements 112 and configured to generate RF energy. The RF generator 114 includes an RF generator controller 116, the RF generator controller 116 configured to control RF energy to the plurality of energy delivery elements 112. The RF generator controller 116 may be implemented using firmware, integrated circuits, and/or software modules that interact or combine together. For example, the RF generator controller 116 may include a memory 118, the memory 118 storing computer readable instructions/code 120 for execution by a processor 122 (e.g., a microprocessor) to perform aspects of the embodiments discussed herein.

The system 100 may also include a computing device 124 (e.g., a personal computer) having one or more controllers. Although a number of different controllers are shown in fig. 1 and described below, the functionality of the various controllers may be implemented in fewer or more controllers (e.g., in multiple modules of a single controller) and/or multiple computing devices.

The computing device 124 of fig. 1 is shown having a display controller 126, the display controller 126 configured to communicate with various components of the system 100 and generate a Graphical User Interface (GUI) to be displayed via a display 128 (e.g., a computer monitor, television, mobile device screen). The display controller 126 may be implemented using firmware, integrated circuits, and/or software modules that interact or combine together. For example, the display controller 126 may include a memory 130, the memory 130 storing computer readable instructions/code 132 for execution by one or more processors 134 (e.g., microprocessors) to perform aspects of embodiments of the methods discussed herein.

The computing device 124 may also include a Graphics Processing Unit (GPU)136 configured to communicate with various components of the system 100. The GPU 136 may be implemented using firmware, integrated circuits, and/or software modules that interact or combine together. For example, the GPU 136 may include a memory 138, the memory 138 storing computer readable instructions/code 140 for execution by one or more processors 142 to perform aspects of embodiments of the methods discussed herein. The GPU 136 may be configured to access other memory in the computing device 124.

The various components of system 100 may be communicatively coupled to one another via a communication link 144. In certain embodiments, the communication link 144 may be or include a wired communication link (e.g., serial communication), a wireless communication link, e.g., a short range radio link, such as bluetooth, IEEE 802.11, proprietary wireless protocols, and so forth. The term "communication link" may refer to the ability to communicate some type of information in at least one direction between at least two components, and may be a persistent communication link, an intermittent communication link, an ad-hoc communication link, or the like. Communication link 144 may refer to direct communication between components and/or indirect communication traveling between components via at least one other device (e.g., a relay, router, hub).

In embodiments, the memory 118, 130, 140 includes computer-readable storage media in the form of volatile and/or nonvolatile memory and may be removable, non-removable, or a combination thereof. Examples of media include Random Access Memory (RAM), Read Only Memory (ROM), Electrically Erasable Programmable Read Only Memory (EEPROM), flash memory, and/or any other non-transitory storage medium that may be used to store information and that may be accessed by a computing device. In certain embodiments, the ablation catheter 102 includes a memory that stores information unique to the ablation catheter 102 (e.g., catheter ID, manufacturer). This information may be accessed and correlated with data (e.g., patient data, ablation parameters) collected as part of an ablation procedure.

Computer- executable instructions 120, 132, and 142 may include, for example, computer code, machine-useable instructions, or the like, such as program components that are executable by one or more processors 122, 134, and 142. Some or all of the functionality contemplated herein may be implemented in hardware and/or firmware.

In certain embodiments, the RF generator 114 and the computing device 124 are separate components housed in a single console 146.

Fig. 2 shows an ablation catheter 200 that may be used in the system 100. The ablation catheter 200 includes an expandable member 202 and a plurality of energy delivery elements 204 (hereinafter referred to as ablation electrodes) secured to the expandable member 202. Ablation electrode 204 is configured and positioned to deliver ablation energy to tissue when expandable member 202 is expanded. As shown in fig. 2, in certain embodiments, the ablation electrodes 204 are arranged in two rows, a set of proximal ablation electrodes 204 and a row of distal ablation electrodes 204. Each of the ablation electrodes 204 is individually addressable and/or may be used with any other ablation electrode 204. The ablation electrode 204 may be operated in a monopolar mode or a bipolar mode. The sets of ablation electrodes 204 may be selected such that the lesion is linear, a dot, a hollow circle, or the like. In embodiments utilizing a monopolar mode, the system 100 may include a return pad.

The ablation catheter 200 includes a visualization system 206, the visualization system 206 including a camera assembly 208 and an illumination source (e.g., a Light Emitting Diode (LED)) disposed on a guidewire shaft 210. The camera assembly 208 may include a plurality of cameras disposed at an angle relative to a longitudinal axis of the ablation catheter 200. The camera is configured to enable real-time imaging (e.g., video) of the ablation procedure within the expandable member 202, including visualization of the expandable member 202, ablation electrodes 204, and cardiac tissue and lesion formation during the ablation procedure. The illumination source provides illumination to the camera to visualize the ablation procedure. In some examples, the surface of the ablation electrode 204 facing the camera is darkly colored (e.g., blackened) to help make it easier to identify the ablation electrode 204 in the image (described in more detail below).

As mentioned above, the computing device 124 of the system 100 includes a display controller 126, the display controller 126 configured to communicate with various components of the system 100 and generate a GUI for display via the display 128. Fig. 3-4 illustrate a GUI and various features and views thereof that may be used in system 100 and displayed via display 128. A user may interact with the GUI (e.g., select icons, enter data) using one or more input devices (e.g., mouse, keyboard, touch screen). The various icons described below may take the form of selectable buttons, indicators, images, etc. on the GUI.

Fig. 3 shows a GUI300 that includes a first region 302 and a second region 304. The first region 302 displays a graphical representation 306 of the electrodes of the ablation catheter, and the second region 304 displays an image 308 (e.g., real-time video) from the ablation catheter. The first region 302 and the second region 304 are shown positioned side-by-side and are circular regions. In certain embodiments, the first area 302 and the second area 304 are separate windows within the GUI 300.

The graphical representation 306 includes a separate electrode icon 310 for each of a plurality of ablation electrodes of the ablation catheter. In some embodiments, each electrode icon 310 is shaped similar to the actual shape of the corresponding electrode on the ablation catheter. Each electrode icon 310 may include a unique numerical indicator 312. For example, the ablation catheter represented by graphical representation 306 includes twelve ablation electrodes in an outer ring and six ablation electrodes in an inner ring, and each of electrode icons 310 is assigned an integer number (e.g., 1 to 18). The user may select or deselect the electrode icon 310 to highlight or de-highlight, respectively, the electrode icon 310 on the GUI 300. As will be discussed in more detail below, in certain embodiments, the computing device 124 provides an initial recommendation of which electrodes (and therefore the electrode icons 310) to activate for a desired ablation path. The initial recommendation may take the form of highlighting certain electrode icons 310 in the GUI 300.

Through the GUI300, the selected electrode icon 310 will be designated as an active electrode (e.g., a source electrode or a sink electrode). Fig. 4 shows an example graphical representation 306 of electrode icons 310, some of which (i.e., electrodes 1-6, 12, and 16-18) are selected as active electrodes, while the remainder of the electrode icons 310 are not selected such that the ablation electrodes associated with the electrode icons 310 will not be active during an ablation procedure.

The displayed real-time video 308 allows visualization of the ablation process. The displayed real-time video 308 may include displaying a video (e.g., a series of images) recorded by one or more cameras. For example, if an ablation catheter (e.g., ablation catheter 200 of fig. 2) includes four cameras, real-time video 308 may display video recorded by each of the four cameras. In such embodiments, the real-time video 308 may display each of four fields of view from the camera that overlap with at least one other field of view.

The GUI300 includes a plurality of icons (e.g., buttons, images, combinations thereof) that are associated with the ablation catheter and the GUI300 itself and that can be used to control or monitor various aspects of the ablation catheter and the GUI300 itself.

Fig. 3 shows a GUI300, the GUI300 including an icon associated with a graphical representation 306 positioned in or near a first area 302 alongside the graphical representation 306. For example, the GUI300 includes three icons (i.e., an electrode selection icon 314, an electrode refresh icon 316, and a source-sink inversion icon 318) that are positioned next to the graphical representation 306 and that affect features of the ablation system 100. Electrode selection icon 314 may be used to select a pattern from a predetermined menu of patterns of electrode icon selection 310 (e.g., inner ring, outer ring of electrode icons 310, all electrode icons 310, none). As described above and in more detail below, in certain embodiments, the computing device 124 provides an initial recommendation of which electrodes (and thus the electrode icons 310) to activate for a desired ablation path. Once the pattern is selected, the selected electrode icon 310 may be highlighted on the GUI 300. Electrode refresh icon 316 may be used to deselect any electrode icons 310 that have been selected. The source-sink inversion icon 318 may be used in bipolar mode to invert which electrode icons 310 correspond to sinks and which electrode icons 310 correspond to sources.

Fig. 3 shows a GUI300 that includes icons associated with real-time videos 308 positioned in or near a second area 304 alongside the real-time videos 308. For example, the GUI300 includes three icons (i.e., a contrast icon 320, a brightness icon 322, and a video refresh icon 324) that are positioned next to the real-time video 308 and affect the characteristics of the real-time video 308. The contrast icon 320 may be used to increase or decrease the contrast of the real-time video 308. The brightness icon 322 may be used to modify the illumination power of the illumination source in the ablation catheter. The video refresh icon 324 can be used to refresh the video feed and/or the display controller if the real-time video 308 encounters a problem.

The GUI300 includes a functional area (ribbon)326 having various icons associated with the ablation catheter and/or the GUI300 itself. Functional area 326 includes a status icon 328 indicating the status of the system and a sonic/sweep icon 330 that initiates a routine for activating an ultrasonic source and for sweeping electrodes on an ablation catheter to identify potentially faulty electrodes. For example, the ablation catheter may be placed in a bath coupled to an ultrasound source, and once the sonic/sweep icon 330 is selected, the routine may turn on the ultrasound source for a predetermined time to remove air bubbles adhering to the ablation catheter prior to the treatment procedure. After expiration of the predetermined time period, the routine may sequentially activate all of the electrodes to determine if any of the electrodes or the RF amplifier are defective. If the ultrasonic source is not connected, the sonic/sweep icon 330 will only initiate the sweep portion of the program.

The functional area 326 further includes: an in-vivo icon 332, the in-vivo icon 332 being selectable to indicate that an ablation catheter has been placed in the patient; an anatomical icon 334, which anatomical icon 334 can be used to identify the pulmonary vein to be treated (e.g., upper right, lower right, upper left, lower left); a power icon 336, the power icon 336 displaying and allowing a user to modify, via arrow buttons, the power level at which a selected ablation electrode is to be activated; a program timing icon 338, the program timing icon 338 displaying and allowing a user to modify, via arrow buttons, the length of time that a selected ablation electrode will be activated; an irrigation flow rate icon 340, the irrigation flow rate icon 340 may be used to control the flow rate of irrigation fluid through the ablation catheter; and a fluid volume icon 342, the fluid volume icon 342 indicating the amount of fluid from within the body icon 332 that was selected to pass through the ablation catheter. Once various selections are made through the icon, data associated with the selections may be stored in memory and/or transmitted to a computing device (e.g., computing device 124 of fig. 1). For example, once the flow rate is selected, the selected flow rate may be sent to the computing device 124 to control the irrigation fluid pump.

As described above, GUI300 allows a user to select which electrodes on an ablation catheter will be active (e.g., source or sink electrodes) via electrode icon 310. The selected and highlighted electrode icon 310 indicates that, if an ablation procedure is initiated, only the ablation electrodes corresponding to the selected and highlighted electrode icon 310 will be active during the ablation procedure. Fig. 3 shows a GUI300 that includes an ablation activation/deactivation icon 344 with text stating "assign" or one or more similar terms. Once the "assign" ablation activation/deactivation icon 344 has been selected, the "assign" text for the ablation activation/deactivation icon 344 is replaced with "ablate," "start," or one or more similar terms.

When the ablation activation/deactivation icon 344 indicates "assign" and is selected, the computing device 124 assigns or designates the selected electrode icon 310 as being active and being a source or sink. For example, FIG. 4 shows the selected electrode icon 310 as either a source electrode or a sink electrode. In fig. 4, selected electrode icons 346 numbered "12", "2", "4", "6", and "17" are indicated as source electrodes, while other selected electrode icons 348 (i.e., those numbered "1", "3", "5", "16", and "18") are indicated as sink electrodes. The unselected electrode icons in fig. 4 (i.e., those electrode icons numbered "7", "8", "9", "10", "11", "13", "14", and "15") are shown without power icons (discussed below) and impedance values (also discussed below). In some embodiments, the source electrodes are highlighted in a different color on GUI300, or are otherwise shown as distinct from the sink electrodes. Further, the unselected electrode icons may not be highlighted or dimmed to further visually distinguish the selected electrode icons from the unselected electrode icons.

Fig. 4 shows that each of source electrode icon 346 and sink electrode icon 348 is associated with an impedance value in ohms (e.g., electrode icon "12" indicates 156 ohms and electrode icon "1" indicates 151 ohms). As will be discussed further below, the impedance value may be an input for determining a recommended ablation path.

Fig. 4 shows each of source electrode icon 346 and sink electrode icon 348, including icons indicating electrical units associated with a given electrode (e.g., power icons). For simplicity, the following description uses power in watts as an exemplary electrical unit displayed and modified via GUI300, but other electrical units (e.g., various forms of electrical energy) may be displayed and modified in place of power. Fig. 4 illustrates a source electrode icon 346 including a source power icon 350 showing the power currently allocated to the corresponding ablation electrode (e.g., 8 watts). Each source power icon 350 may be selected to display a power selector icon to increase or decrease the power associated with the respective source electrode. Fig. 4 also shows each of the sink electrodes 348 including a power estimation icon 362. Each of the power estimation icons 362 displays an estimated power associated with a given sink electrode 348. The estimated power displayed in the power estimation icon 362 is based on, for example, the power allocated to each of the source electrodes 346 and the distance between a given sink electrode 348 and the source electrodes 346. For example, the source electrode 346 will divide its energy between the sink electrodes 348, but more of its energy will be delivered to the sink electrode 348 that is positioned closer to a given source electrode 346.

As described above, the computing device 124 may be configured to provide initial suggestions of which ablation electrodes 204 (and thus which electrode icons 310 to initially highlight) are to be activated for the ablation procedure. The recommended path helps to enhance the process of determining which electrodes to activate for a given ablation procedure. For example, the recommended path may make it easier and/or faster for the physician to select which ablation electrodes 204 to activate for the ablation procedure.

The process 400, 500 of recommending a path is schematically illustrated in fig. 5 and outlined in fig. 7. The analyses and steps shown in fig. 5 and 7 and described below may be performed in parallel or sequentially in various orders. Also, in some embodiments, not all of the analysis and steps need be performed to recommend a path. The processes 400, 500 include identifying which ablation electrodes 204 sufficiently contact tissue through visual analysis 402 (e.g., analyzing images collected by one or more cameras on the ablation catheter 200) and/or impedance analysis 404 (e.g., analyzing impedance measurements of the ablation electrodes 204 on the ablation catheter 200).

As mentioned above, one or more cameras positioned on the ablation catheter 200 record images for display in the GUI 300. Example images 406A-406C are shown in fig. 6A-6C. Images 406A-406C that may be taken over multiple cardiac cycles may be input to the computing device 124 and analyzed to determine which of the ablation electrodes 204 are likely to be in sufficient contact with tissue. The visual analysis 402 may include first identifying or recognizing the ablation electrodes 204 in the images 406A-406C (step 502 in fig. 7). For example, the images 406A-406C may be analyzed by the neural network 408 (described in more detail below) to determine the location and boundaries of any ablation electrodes 204 in the images 406A-406C. As shown in fig. 6A-6C, each of the ablation electrodes 202 in the images 406A-406C is surrounded by a respective frame 450A or 450B. These blocks 450A and 450B visually indicate that the neural network 408 has identified or recognized the ablation electrode 204 in the images 406A-406C.

Once the ablation electrodes 204 in the images 406A-406C are identified, the visual analysis 402 may include determining whether the identified ablation electrodes 204 are sufficiently in contact with tissue (step 504 in fig. 7). In certain embodiments, the neural network 408 determines the likelihood of contact by comparing the contrast of pixels in the images 406A-406C around the identified ablation electrode 204. For example, pixels around the ablation electrodes 204 associated with lighter colors may indicate contact with tissue, while pixels associated with darker colors may indicate the presence of blood between one of the ablation electrodes 204 and the tissue. In fig. 6A-6C, the boxes associated with reference numeral 450A are those boxes for which the neural network 408 has determined to include ablation electrodes 204 that are likely to be in contact with tissue, while other boxes 450B indicate ablation electrodes 204 that are unlikely to be in contact with tissue.

In certain embodiments, the output of the visual analysis 402 is a list of ablation electrodes that can be selected for activation based on which of the ablation electrodes 204 are likely to be in sufficient contact with tissue. Further, each ablation electrode 204 included in the list may be associated with a contact level or confidence level of the contact. For example, the visual analysis 402 may indicate that certain ablation electrodes 204 may have better tissue contact than other ablation electrodes 204 (such as ablation electrodes 204 surrounded by pixels having lighter colors).

In general, the neural network 408 is a computational model based on the structure and function of a biological neural network. The neural network 402 may be implemented under a variety of methods, including a Convolutional Neural Network (CNN) method, and the like. CNN evaluates data (e.g., images) in the form of a plurality of arrays, dividing the data into a series of stages and examining the data for learned features. Fig. 8 is a simplified visual representation of an example implementation of image analysis CNN 600. CNN 600 (and thus neural network 402) may include additional features (e.g., layers).

The image 602 is input to the CNN 600, which CNN 600 extracts the image 602 in a first convolution layer 604 to identify the learned features. In the second convolutional layer 606, the image 602 is transformed into a plurality of images, with the learned features each emphasized in a corresponding sub-image 608. The image sub-image 608 is also processed to focus on the feature of interest, and further processing isolates the portion 510 of the image that includes the feature of interest. Output layer 612 of CNN 500 receives values from the last non-output layer and classifies the feature of interest based on the data received from the last non-output layer.

The neural network 408 may be "trained" using supervised or unsupervised methods. For example, a set of "training data" (e.g., known inputs and known outputs) may be used to train the neural network 408. Using supervised methods, the input training data may be images collected from the ablation catheter that are classified with output data (e.g., presence/boundaries of ablation electrodes; an indication of whether the ablation electrodes appear to be in contact with tissue). The known inputs and outputs are fed to an untrained neural network or a partially trained neural network (e.g., a neural network trained to recognize objects but not necessarily ablation electrodes) that processes this data to train itself to parse/compute the results of additional sets of data with new inputs and unknown outputs. Using an unsupervised approach, the input training data may similarly be images collected from the ablation catheter that are not manually or semi-automatically classified but are compared to the actual lesion features produced by the activated ablation electrodes in the images. As a result, under either supervised or unsupervised approaches, the trained neural network can predict an output (e.g., an indication of the presence/boundary of an ablation electrode and whether the ablation electrode appears to contact tissue) from a new set of inputs (e.g., a new image obtained from an ablation catheter).

The process 400/500 also includes an impedance analysis 404 in which the impedance values measured by the ablation electrodes 204 are analyzed (step 506 in fig. 7). The impedance values may be collected and input to the computing device 124. For example, the GUI300 may include a button that initiates a routine that involves activating the ablation electrodes 204 for a period of time to determine the impedance value measured by each ablation electrode 204. The computing device 124 may compare the measured impedance values to a predetermined threshold range to determine which ablation electrodes 204 may be sufficiently contacting tissue. For example, if the predetermined threshold range is 50 to 250 ohms, an ablation electrode 204 measuring an impedance value below 50 ohms or above 250 ohms will be determined to not be in sufficient contact with tissue. The computing device 124 determines a list of ablation electrodes 204 that measure impedance values within a predetermined threshold range.

The computing device 124 uses the outputs of the vision-based analysis and the impedance-based analysis to determine a list 410 of ablation electrodes 204 that may be selected for activation (step 508 in fig. 7). In some embodiments, the selectable ablation electrode must be determined to be in sufficient contact with tissue under both the vision-based analysis 402 and the impedance-based analysis 404. For example, if the vision-based analysis 402 determines that a given ablation electrode 204 is likely to contact tissue, but that same ablation electrode 204 is associated with an impedance outside of a predetermined threshold range, then that ablation electrode 204 will be determined to be non-selectable. In some embodiments, additional analysis may be used to generate a list of selectable ablation electrodes 204. For example, if one of the ablation electrodes 204 is determined to be defective during the insonification routine, that ablation electrode 204 may be identified as being non-selectable. The list of selectable electrodes 410 is used as an input to an ablation path analysis 412, which is used by the computing device 124 to determine a recommended ablation path (step 510 in fig. 7).

Because not all theoretically possible paths are valid or desired paths, the computing device 124 may apply various rules and/or preferences to eliminate and/or give more or less weight to certain paths as part of determining a recommended ablation path. An example rule is that two adjacent electrodes cannot both be sources, so when applying this rule, paths with adjacent sources are excluded from consideration. Another example rule is that a sink cannot be associated with an estimated power value above a predetermined threshold, so when this rule is applied, paths that result in excessive power at one or more of the ablation electrodes 204 are excluded from consideration. For paths that are not excluded, certain preferences may be applied to give more or less weight to paths with preferred characteristics. An example preference is that sinks in a given path will have similar (e.g., balanced) estimated power values, thus resulting in paths of similar estimated power values being weighted higher than other paths. Another example preference is that the recommended ablation path is "closed" (i.e., forming a closed circuit having a circumferential shape) rather than "open," so that such path is weighted higher than other paths. Another example preference may include weighting the vision-based analysis differently than the impedance-based analysis. As mentioned above, the visual analysis may include a value indicative of the estimated contact level of each selectable ablation electrode 204. In ablation path analysis 412, ablation electrodes 204 associated with higher contact levels may be given higher preference.

Given the constraints and preferences, the computing device 124 may determine a recommended ablation path. In certain embodiments, the computing device 124 applies graph-theoretic methods to determine the recommended ablation path. Applying graph theory, the ablation electrodes may be represented as nodes (or vertices) in the model connected by edges (e.g., undirected edges). For example, fig. 9 shows a diagram 700 of ablation electrodes 204 connected by edges 702 (only some of which are associated with reference numeral 702 in fig. 9). Like the electrode icons 310 in the GUI300, each of the ablation electrodes 204 is represented and shown as having a unique numerical identifier (e.g., "1", "2" through "18"). Applying the graph theory, constraints, and preferences, the computing device 124 may determine a recommended ablation path. In some examples, the computing device 124 may apply another trained neural network (e.g., using a multi-layer feed-forward network approach, a recurrent neural network approach) to determine the recommended ablation path.

The recommended ablation path from the ablation path analysis 412 may be communicated to the display controller 126. The display controller 126 may cause certain electrode icons 310 to become highlighted in the GUI300 as shown in fig. 4. The highlighted electrode icons 310 are those associated with the ablation electrodes 204 within the recommended ablation path. Further, the computing device 124 may automatically assign which ablation electrodes are sources and which are sinks, and assign the recommended power values to the sources.

The recommended ablation path and associated parameters (e.g., source, sink, power) may be modified by the user via GUI 300. Once the ablation path and associated parameters are established, the user may select an ablation activation/deactivation icon 344 (see fig. 3). Selecting the ablation activation/deactivation icon 344 initiates and/or ceases energy delivery to the ablation electrode 204 of the ablation catheter 200. Once the activation/deactivation icon 344 is pressed to initiate energy delivery, the graphics in the activation/deactivation icon 344 change (e.g., to a stop symbol). Further, once the activation/deactivation icon 344 is selected to initiate energy delivery, a signal is transmitted to an RF generator (e.g., RF generator 114 of fig. 1) and/or an RF generator controller (e.g., RF generator controller 116 of fig. 1) to begin delivering energy to the selected ablation electrode 204 of the ablation catheter 200.

Various modifications and additions may be made to the exemplary embodiments discussed without departing from the scope of the present invention. For example, although the embodiments described above refer to particular features, the scope of the invention also includes embodiments having different combinations of features and embodiments that do not include all of the described features. Accordingly, the scope of the present invention is intended to embrace all such alternatives, modifications and variances which fall within the scope of the appended claims and all equivalents thereof.

Claims (15)

1. A computer-implemented method, comprising:

recommending, via the one or more processors and the graphical user interface, a set of ablation electrodes to activate based at least in part on an analysis of an image containing the ablation electrodes.

2. The method of claim 1, wherein the analyzing comprises identifying the ablation electrode in the image.

3. The method of any one of claims 1 or 2, wherein the analyzing comprises comparing contrast of pixels in the image around the ablation electrode.

4. The method of any of claims 1-3, wherein the analysis determines which of the ablation electrodes are selectable.

5. The method of any one of claims 1 to 4, wherein the analyzing is performed by a trained neural network.

6. The method of any of claims 1-5, wherein the recommendation is further based at least in part on an impedance measurement.

7. The method of claim 6, wherein the impedance measurements determine which of the ablation electrodes are selectable.

8. The method of claim 7, wherein the selectable ablation electrodes are ablation electrodes associated with impedance measurements within a predetermined range of impedance values.

9. The method of any of claims 4, 5, 7, and 8, further comprising:

assigning, via the one or more processors, each of the recommended ablation electrodes as a sink or a source.

10. The method of any of claims 1 to 9, wherein the recommendation is based at least in part on balancing estimated power received by a source electrode.

11. The method of any of claims 1 to 10, wherein the recommendation is based at least in part on limiting an estimated power received by a source electrode below a predetermined threshold.

12. The method of any one of claims 1 to 11, wherein the recommended set of ablation electrode sets forms a closed circuit.

13. A computing device adapted to perform the steps of the method of claims 1 to 12.

14. A computer program product comprising instructions for causing one or more processors to perform the steps of the method according to claims 1 to 12.

15. A computer readable medium having stored thereon the computer program product of claim 14.

Images